PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
Commits on main
fix: support aggregated Granite MoE tensors in gguf export (#2175)
5a24c84 Heng Guo committed 3 months ago
fix DiffusionCompressionOrchestrator missing nsamples (#2179)
f239f8c Wang, Chang committed 3 months ago
Add support for kimi_k25 INT4 packed source models and related tests (#2156)
6743779 Xin He committed 3 months ago
Add timeout decorators to improve test case stability (#2155)
cee7109 Sun, Xuehao committed 3 months ago
fix: prevent quantized layers from being reinitialized by Transformers (#2147)
a5e7393 Heng Guo committed 3 months ago
Enhance unit tests with timeout handling and stability improvements (#2135)
e4a02cc Sun, Xuehao committed 3 months ago
[fix] XPU crash due to circle call of nonzero() and index_select() (#2117)
2800369 Xin He committed 3 months ago
Enable torch.compile for selected tests and update related configurations (#2124)
a2f6b31 Xin He committed 3 months ago
support fused MoE quantization and export for GGUF format (#2072)
1fca0c7 Heng Guo committed 3 months ago